HIV antiretroviral (ARV) resistance, substance abuse, and ARV diversion are significant challenges to the global rollout of ARV treatment. They are associated with increased morbidity, mortality, and treatment cost. Our group is among the first to report an emerging form of substance abuse at the intersection of these phenomenon: the recreational use of ARVs. Our formative work in Durban, South Africa reveals that a highly addictive drug cocktail known as whoonga likely contains diverted ARVs, especially efavirenz. Smoked by adolescents and adults, our research demonstrates that whoonga use is highly stigmatized and widely recognized to confer significant risks to the health and safety of users and their communities. Additional study is needed, however, to determine the context of recreational ARV use and the extent to which whoonga use is associated with exposure to ARVs. Because HIV prevalence in this area is among the highest in the world, any exposure to ARVs from recreational use could result in the development of ARV-resistant HIV strains. Overview of Project. We propose to examine recreational ARV use in the context of whoonga use among individuals admitted for inpatient substance abuse treatment at Durban's largest substance abuse treatment center (SANCA Durban) and substance users from their social networks. Preliminary data from our group shows that whoonga is the primary drug of abuse for approximately one-third of individuals seeking inpatient substance abuse treatment at SANCA Durban. Over an 18-month period, we will survey the substance abuse patterns of whoonga users (N=100) and nonusers (N=100) admitted to SANCA Durban and an additional 200 men and women with active substance abuse problems recruited from their social networks. We will match self-report of illicit drug and/or whoonga use against results of a toxicology screening using a novel application of high-resolution accurate mass spectrometry to detect illicit drugs and ARVs originally developed by the Toxicology Core of the HIV Prevention Trials Network Laboratory. We will thereby assess ARV exposure from whoonga use in this group, including among HIV-infected and HIV-uninfected individuals who deny any exposure to ARVs other than from recreational use. To complement this quantitative work and collect the data needed for an R01 study, we will also conduct semi-structured qualitative interviews among whoonga users from SANCA Durban and their networks (N=45 total; 15 SANCA participants, 15 male/15 female network participants) and key informants (N=10) to learn more about whoonga and whoonga users, whoonga-related ARV diversion, and relevant community-based capacities to address this phenomenon. Combining these approaches, we will document whether ARVs are used recreationally, gain important insight into both the whoonga phenomenon and any relationship among ARVs, HIV, and this new drug cocktail, determine if whoonga use results in detectable ARV exposure, and identify strategies for further study and intervention.